Two complementary studies aimed at the recently unemployed are to be run in near parallel using shared measures and conceptual frameworks. One, by the NIMH-funded Michigan Prevention Research Center (MPRC), is a recently- initiated second-generation experiment in the enhancement of self-efficacy and peer support for coping with job loss. The second is the project proposed here: a longitudinal study of couples transactions that support or undermine job-seeking and mental health. The two research teams intend to combine the results of these efforts and collaborate in 1994 on a third generation of field trials on preventing mental illness -- testing interventions that bolster self-efficacy on the job-seeker with both peer and spouse support. Although considerable epidemiological evidence indicates the clear importance of the marital relationship in preventing mental illness, the important components of that relationship and their antecedents are not well-enough understood to ensure successful prevention programs. The proposed study will identify those patterns of spouse interaction that maintain motivation in the job-seeker and support positive mental health in both partners. It will also identify proximal and distal antecedents of these interaction patterns, including role expectations, mood state, and attributions concerning problem cause and problem solution. The general intent is to identify the most promising behavioral, cognitive, and affective targets for preventive intervention in the next generation of field trials to promote mental health and effective coping among the recently unemployed. A longitudinal three-wave panel design will be used to follow 250 recently unemployed adults and their spouses/partners selected to represent a broad distribution on demographic characteristics including gender, age, race, education, and occupation. There will be an initial measurement shortly after job loss (T1) and resurveys at 4 and 12 months (T2 and T3). In-home interviews at T1 will include videotaping of couples interactions. Both partners will be interviewed by telephone on four days after T1 and T2 to generate daily diary data. Diary data will be used to examine the daily variation in constructs and to test for short-term causal effects. The project will also include followup interviews after T3 of outliers, or couples who are consistently over- or under-predicted by our models across T1 through T3. The findings will be used to generate spouse interventions aimed at increasing job-seeker motivation and decreasing depression; these interventions will be pilot tested at the end of the project. The final products of the study will be the findings, resultant measures, and theory- and result-driven proposals for a new generation of prevention experiments that combine enhancement of self-efficacy with peer and spouse support.